A neural network-based scheme for decision directed edge-adaptive Kalman fi
ltering is introduced in this work. A backpropagation neural network makes
the decisions about the orientation of the edges based on the information i
n a window centered at the current pixel being processed. Then based upon t
he neural network output an appropriate image model which closely matches t
he local statistics of the image is chosen for the Kalman filter. This prev
ents the oversmoothing of the edges, which would have otherwise been caused
by the standard Kalman filter. Simulation results are presented which show
the effectiveness of the proposed scheme.